Phantom Bridge Effects

[Brian Buchanan]

Hi all,

We’ve been conducting some 2D unsteady flow analyses that incorporate bridges into the 2D flow area. When adding bridges, I noted some small but odd backwater effects happening at flows that were well below (5ft below in some cases) the lower chord of the bridge decks (note all modeled bridges were channel-spanning and had no piers or abutments). I found this strange so I installed the HEC example datasets and ran the Bald Eagle Creek Example. First, I ran the “Single 2D area with Bridges FEQ” plan with the only modification that I shortened the model run time and chose diffusive wave as the equation set to speed things up. Then I simply removed the “Highway 150 Lowe” bridge and re-ran the model. Looking at the “Connection Data Editor”, it was apparent that the lower chord of the Highway 150 Lowe bridge deck was 560.5ft. When we examined a longitudinal profile of the WSE at the Highway 150 Lowe bridge with and without the bridge in place, we see that WSE is roughly at 556ft for both scenarios (well below the bridge deck)…..and yet the WSE profiles are NOT the same (i.e. deleting the bridge removes the small upstream hydraulic effect). I have run numerous models with different terrains, different bridge dimensions, bridge modeling approaches and settings, HTab parameters, “Min Weir Flow El” etc. In all cases this phantom bridge effect is there. Any thoughts or advice? Many thanks, Brian

[Zain]

Hi Brian. I am working on a project that requires modeling quite a few bridges in 2D. I believe I am also experiencing the same anomaly as you are.

The WSE is well below the low chord and its completely an open channel flow. There is an exagerrated drop in WSE (HW and TW) at the Bridge 2D Connection CL. For me, this exegerration becomes a little okay when the flows become large but for low flows, the profile is very weird. When I remove the bridge and run the 2D, the WSE does not has that drawdown (which is logically correct). I have modelled this bridge in 1D also and for that, I do not have that weird drop but inputting bridges in 2D behaves abnormally. After reading your post, i am also checking the Bald Eagle Creek example.

Are you able to figure out this bridge problem?

Thank you.

Nayyar

[Chris Goodell]

Keep in mind that when you add in a bridge, it will use whatever low flow bridge method you select to add an extra force term to the cell faces at the bridge centerline. So there will be some difference for sure, even if the water does not touch the bridge deck. You might be able to minimize this effect if you use very small n values on the bounding “dummy” cross sections, but that is just a guess. My suggestion would be to not include the bridge if the water doesn’t touch the deck and there are no piers.

[Zain]

Noted Chris. Thankyou for your response.

[Author]

Posts